Every 1 in 59 children is diagnosed with Autism Spectrum Disorder (ASD) every year. With an estimated 1.4 million cases registered in the year 2018 itself. These children often struggle to recognise facial expressions and are consequently hindered in understanding social cues. Overcoming such problems requires an intensive behavioural therapy which is not only expensive, but also not universally accessible and consistently administered.
What it does
We have therefore decided to develop a system to automate facial expression recognition that runs on a wearable cap and delivers real-time visual (auditory) social cues to the autistic children, with the goal of creating a behavioural aid for these children that maximises behavioural feedback and provides an immersive memorisation, while minimising the distractions to the child.
How I built it
raspberry pi, Movidius Neural Compute Stick, camera, caps - openCV, Python
Challenges I ran into
lack of friendship with time, limited opportunity to explore Bremen, no cultural events and opera.
Accomplishments that I'm proud of
using Movidius for the first time, learning about El Salvador, and about the secret recipe of Pupusa
What I learned
have openCV installed before, GitHub commands, Raspi management
What's next for CuedIn
Complete 2.0 Version with a targeted Amazon partnership